POLICY MEMO

A Brief History of Internet Regulation

By Ev ehrlich March 2014

Executive Summary From the beginning of “community antenna Proposals to regulate the Internet are often TV” through the 1990s, a parallel but more presented as “new” solutions to deal with limited effort was made to regulate the nascent modern problems, but the most significant of cable industry. While these regulations had these proposals, such as “network neutrality” some success, technological change quickly and common carrier rules on unbundling outstripped them—both in the telephone business and interconnection, are actually vestiges and the emerging field of high-speed data—and of long-outmoded ways of thinking about a bipartisan consensus formed in the early 1990s policy. This paper explores that additional steps were needed to promote the relevant regulatory history, offering critical competition in all these arenas. context to today’s Internet policy debates. The result was the Telecommunications Act of From the early days of the AT&T monopoly well 1996, watershed legislation that marked the end into the 1990s, regulators, the courts and the of the telephone age and the beginning of the Congress engaged in a lengthy effort to protect Internet age from a policy perspective. The Act consumers and ultimately bring competition embraced and codified the FCC’s distinction into the markets for local and long-distance between traditional telephony/telecommunications telephone service. This included strict “common services and the emerging world of information carrier” utility regulations and mandatory services, with strict common carrier rules limited interconnection requirements and ultimately to the former. On the telephone side, this meant the 1984 Modified Final Judgment, which forced a stifling regime of mandatory “unbundling” and the breakup of AT&T into regional Baby Bells. rigid price controls, while giving the private

About the author Ev Ehrlich is a senior fellow at the Progressive Policy Institute and the president of ESC Company. Policy Memo Progressive Policy Institute

sector more latitude to innovate and invest on Ultimately, three key lessons emerge from this the “information services” side. The 1996 Act may policy review. First, information services and not have specifically contemplated the rise of the telecommunications services really are different, broadband Internet (the idea of an “information and broadband has flourished as an information superhighway” was in the air, but the exact form service free from ill-fitting and stifling common it would take was still unclear as a matter of carrier constraints. Second, investment and both technology and policy), but by protecting capital flow to where regulation (or the absence information services from the common carrier thereof) encourages them to flow. And third, framework, the Act set the stage for the dynamic technology, business models, and consumer growth we have seen in American broadband. behaviors change and, as they change, the meaning and effect of different regulatory The result was a boom in cable broadband proposals change as well. investment that telecommunications providers attempted to counter by offering DSL services. Introduction But any new DSL capability they constructed An active public policy debate is underway had to be leased out to competitors at below regarding Internet regulation. At its core lie market prices under the unbundling regime, a series of proposals that address what some which limited their efforts. When fiber and DSL advocates perceive as an absence of competition were relieved of their unbundling obligation in in the broadband industry. Most important the early 2000s, however, capital poured in and among these are the imposition of “net neutrality” these services flourished as fixed-broadband rules that would compel all Internet traffic competitors to cable. In fact, that competition to travel at the same speed and on the same drew a competitive response from cable, in terms; restrictions on usage-based pricing or turn leading to a virtuous cycle of improvement other broadband business models, including and enhancement resulting in the United prohibitions on creating a “two-sided” market in States ascending to the upper reaches of the which Internet-based business might pay to reach International broadband rankings. network users (much as advertisers pay to reach newspaper readers); mandatory interconnection This background sheds important light on current rules that would empower regulators to dictate calls to impose “new” regulations on broadband prices or procedures for the exchange of data either through “network neutrality” rules or by among networks in the Internet’s labyrinth; limits reclassifying it as a “telecommunications service” on the participation of selected firms in auctions subject to common carrier obligations. While of electromagnetic spectrum; and facilities advocates suggest otherwise, these proposals are “unbundling,” which would require providers of clearly not new, but would represent a return Internet infrastructure to make portions of their to the dated—and in the view of this paper physical networks available to all competitors at failed—approach that the bipartisan 1996 Act was prices set or approved by the government. designed to sweep away. Most of these proposals for network micromanagement, forced sharing Some advocates claim that these “new” ideas are of investments, and government influence on needed to address new problems; they contend pricing have been associated with low investment that Internet service is not provided competitively and innovation. These rules may have made sense or that it is too important to leave unregulated. when the problem was how to protect consumers In a separate paper, we will argue that there is in the days of the sanctioned Ma Bell monopoly, no “competitiveness” problem in the provision but the business and consumer landscape is of broadband when conventional yardsticks such dramatically different today in almost every as investment, innovation, or prices and profits regard. are considered.1 But regardless of the merits of the assertion that the provision of broadband is

2 Policy Memo Progressive Policy Institute

uncompetitive, the regulatory proposals that make Commitment,” under which it would allow non-Bell up this debate are not, in fact, new ideas. Instead, local companies to connect to its interstate (long most of the proposals for Internet regulation are distance) system. It also made other concessions the regulatory tools government applied to the (including divesting Western Union, the second “T” telephone system during its period as a regulated, in its name) in return for a government sanctioned government-sanctioned monopoly decades ago. monopoly.

The deal hardly held AT&T back. The behemoth Most of the proposals for still used access to the long distance market as a Internet regulation are way to disadvantage local competitors. Regulation the regulatory tools of its acquisitions was lax. And, meanwhile, in a different realm, radio was burgeoning. All of this government applied to combined to create pressure for a new federal the telephone system framework towards telephone and radio, which during its period as a led to the Communications Act of 1934 (the “1934 Act”). regulated, government- sanctioned monopoly The 1934 Act established a framework for telephone regulation that would last for half a century. decades ago. AT&T’s interstate long-distance lines would be subject to federal regulation, and intrastate lines But the circumstances then were very different would be regulated by the states. Interconnection from those we face today. For that reason, a among these systems was mandatory and regulated, review of the history of Internet policy, from the and a system of settlements was created to allocate formation and break-up of the to revenues and costs between and within the long- the current-day court cases that address many distance company and local operating companies of the underlying questions regarding the legal and to balance accounts. The 1934 Act created framework for regulation, is in order. Where the Federal Communications Commission (FCC) were these policy proposals first raised, and in to carry out the federal regulatory role in both what context? And what can be learned from our interstate telephony and radio. experience with them? Those are the questions this paper will address. This system survived for decades, but was ultimately undone by technological change, The next section provides a brief history of U.S. beginning in the long distance market. Microwave regulatory policy towards the Internet from technology created new competitors in the long- the perspective of telecommunications.2 That is distance market, led by MCI. The long-distance followed by an attempt to identify the roots of market was particularly ripe for picking, as leading contemporary policy proposals in that regulators generally allowed inflated long-distance history and to apply the lessons learned from this pricing to cross-subsidize , review. emergency response, and other local services, especially residential phone service. AT&T Telecomm Roots fought these competitors in court and through By the 1910s, AT&T had a commanding discrimination in its practices (for example, position in the U.S. long-distance telephone requiring competitor long-distance companies to market, and it used that position to begin use a “dial-in” number to get access), but ultimately, acquiring local companies, often by denying the pressure of technological progress became too others interconnection. In 1913, facing a federal great. In 1982, AT&T and the Justice Department antitrust suit, AT&T entered into the “Kingsbury entered into an antitrust consent decree—the

3 Policy Memo Progressive Policy Institute

“Modification of Final Judgment” (“MFJ”). While did not appear until the 1990s) contributed to a the name may sound like a panel on the ceiling backlash that led to the Cable Television Consumer of the Sistine Chapel, it actually restructured the Protection and Competition Act (“Cable Act”) in telephone market. 1992, which gave the FCC authority to regulate cable prices and mandated that content developed and owned by the cable companies (e.g., HBO at A bipartisan consensus that time) be made available to competitors. The gradually emerged in FCC soon imposed price rollbacks on cable services using its new authority, with disastrous results response to these for investment in both the cable system’s physical developments that steps networks (which limited their ability to emerge as were needed to promote an important source of telephone competition) and cable programming services (which, confusingly, are competition in all of also commonly referred to as “networks”). these arenas. Like the 1934 Act, the MFJ and the Cable Act “worked” so long as their underlying premises The MFJ’s major provisions were, essentially, regarding technology and market structure did not three-fold. First, AT&T would divest itself of local change. But the prospect of greater competition affiliates and become a long distance company in both long-distance and local telephony was (inter-exchange carrier) in a competitive market. clear—long-distance rates were dropping and long- Second, the local telephone companies would be distance providers were beginning to bypass local organized into seven regional competitors that loops in some situations. There was already some would be barred from providing long distance local competition for business communications in and information services, and whose local services local phone markets. Meanwhile, Direct Broadcast would continue to be regulated, predominantly Satellite (“DBS”) entered the market and cable by the states. And, third, the local telephone companies were experimenting with what then companies were barred from manufacturing both seemed to be futuristic systems involving hundreds telecommunications network equipment and of channels and even “high-speed data,” even telephones and other so-called “customer premises as cable investment slowed in response to price equipment,” buttressing the FCC’s earlier decision regulation. to end the Bell System’s monopoly over this hardware. A bipartisan consensus gradually emerged in response to these developments that steps were The regulation of cable proceeded on a parallel needed to promote competition in all of these track. Today we think of telecom and cable arenas—local telephony, long distance, cable, companies as being in the same business. But their and even the prospect of “video dial tone,” that roots were obviously different. Cable was subject is, television delivered by the phone system. This to local regulation, but the profusion of inefficient, consensus culminated in the Telecommunications disparate regulatory standards and procedures led Act of 1996. That Act not only addressed that government and industry alike to embrace the 1984 objective, but—to some extent unwittingly—set the Cable Communications Policy Act, which imposed stage for the broadband Internet of today. some discipline on the process, gave the companies more latitude to price, and fed a cable investment The Telecommunications Act of 1996 boom in that decade and afterwards. The Clinton Administration brought with it a very specific view of the future of telecommunications. But the increasing prices of cable services and It saw the potential for an explosion in then-minimal competition (satellite competitors “information services” and believed strongly

4 Policy Memo Progressive Policy Institute

that relying on private investment and markets But, nonetheless, the 1996 Act set the framework would be the best route to promoting innovation, within which the high-speed Internet arose and raising investment capital, and managing the telephone, cable, and the Internet converged once uncertainties about the shape these future services technological change bridged the divisions among would take. them. The Act did so through two, interrelated features. First, the Act embraced and codified the It applied this perspective widely. For example, it distinction between telecommunications services and argued in international fora that the Internet did information services. The former was essentially not belong in the province of the International the telephone system that had evolved through Telecommunications Union, since it was different the Kingsbury commitment, the 1934 Act, and than standard telephony. It made sure that the the MFJ. The latter, “information services,” would governance structure of the Internet remained soon include the cable and broadband worlds, but in private hands—those of the user community— at that moment mostly consisted of the private rather than being moved to government decision data networks used by larger companies to link makers. It took steps to facilitate Internet computers, faxes, and other gear, and the then- commercialization. And, with the Congress’ fledgling world of dial-up services like Compuserve cooperation, the Omnibus Budget Reconciliation and America Online. Second, the 1996 Act treated Act of 1993 preempted state and local regulation those different classes of service very differently. of entry by and rates for mobile telephone services, Information services were by law excluded from even though these were not yet regarded as direct traditional common carrier regulation, while competitors to other pieces of the telephone traditional telecommunications continued under industry, let alone what we now think of as the this regulatory frame. Cable systems received some broadband Internet. And the deregulation of other limited regulatory relief, such as a sunset on mobile telephony was accompanied by public some pricing regulations by 1999 (although basic auctions of spectrum to support it (championed rates remain regulated today unless the FCC finds and overseen by then-FCC Chairman Reed Hundt), that adequate competition exists in a given market). the first of which took place a year later and which At the same time, the Baby Bells were allowed resulted in increasing the number of competing entry into the cable business, from which they had wireless companies. The burgeoning success previously been barred. Yet broader deregulation of mobile telephony and, subsequently, mobile of telecommunications, while an aspiration, would broadband has its roots in this decision. await the emergence of expanded competition.

But the most important manifestation of the Clinton Administration’s perspective was the The 1996 Act was the Telecommunications Act of 1996 (“the 1996 Act” or watershed event that marked “the Act”). The 1996 Act was the watershed event that marked the end of the telephone age and the the end of the telephone age beginning of the Internet age in the public policy and the beginning of the realm. Internet age in the public Today, we regard the convergence of telephones, policy realm. television, and the Internet as a fact of life. In 1996, it was considered a futuristic proposition. The purpose of the 1996 Act was both to promote The Act attempted to create competition in local the convergence of these different delivery telephone markets in three ways still relevant modes through inter-modal competition—and to today. First, it prohibited states from sanctioning foster competition within each of its individual any local monopoly by the “Baby Bells.” Second, component sectors (intra-modal competition). it required the Baby Bells to interconnect with

5 Policy Memo Progressive Policy Institute

emerging competitors on state supervised rates, The first commercial point-and-click browser was terms, and conditions, and created telephone developed by , and public awareness of number “portability,” allowing users to keep the Internet boomed. Dial-up services such as their number if they switched providers in order AOL and Compuserve went from being “walled to jumpstart local phone competition (a critical gardens” of content to gateways to the Internet. requirement both to get competition started and The beginnings of a race by investors to fund the maintain it in the long run). Third, and most digital Internet and the enterprises that would use controversially, it established that emerging local it began, setting in motion the dot.com bubble competitors to the Baby Bells could get wholesale of the late 1990s. The 1996 Act was signed on connections on the Bells’ existing local phone February 8 of that year, and only 10 months later, networks and then resell that capacity in the retail Fed Chairman soliloquized over market—a procedure known as “unbundling” “” in the stock market. (since it required the phone companies to separate, or “unbundle,” the capacity on their systems that supported voice calling and sell it at a below- The idea of an retail rate). Unbundling spawned a new class of “information “Competitive Local Exchange Carriers” (CLECs) superhighway” had who availed themselves of this privilege and, to a great extent, became the lynchpin of the debate caught on, although it over implementation and enforcement of the Act. was unclear exactly what it would be. As mentioned, the 1996 Act didn’t expressly compel the convergence of phones, cable, satellites, mobile phones, and the broadband Internet. But, at the same time, the sense that we faced So, while the 1996 Act addressed the specifics a dramatically different future was in the air. of both intra-modal and inter-modal competition The idea of an “information superhighway” had for traditional telephone services, it also pointed caught on, although it was unclear exactly what towards a radically different future that it could it would be. The Clinton Administration, and not define but nonetheless sensed. In 1996, the Vice President Al Gore and Commerce Secretary cable and telephone worlds were distinct, and Ron Brown in particular, were enthusiasts for the reciprocal interest in competing was guarded. promise of such innovation—the Administration While some cable companies showed interest was busy at work on various policies to facilitate in telephony, the cable industry as a whole did the Internet’s commercial development. And not race into it, although some began offering important developments were setting the stage voice services after the Act was passed and even as the 1996 Act was being debated and others contemplated investments in circuit- written. switched phone service. Phone companies had experimented with “video dial-tone” trials but As the 1996 Act was being debated, important there was no evidence that there would be technological developments were in motion. The consistent and adequate consumer demand switch from analog transmission of signals (in justifying a permanent commitment. High-speed which a signal such as your voice is sent as a broadband (at least by today’s standards) did not continuous and uninterrupted stream of waves) yet exist in the commercial market, much less the to digital communications (in which a signal residential market. But the Act’s intuition was is repeatedly sampled at very high speeds and that, if the markets for the two types of services converted into 0s and 1s) was proceeding rapidly, were competitive, and if the two systems entered driven by cost and efficiency concerns as well as into facilities-based competition—meaning wholly the potential product offerings it made possible. separate systems would connect the household to

6 Policy Memo Progressive Policy Institute

services—long-term inter-modal competition could be strengthened at the same time. As baseball executive and savant Branch Rickey is said to have remarked, “Luck is the residue of design.” Or, as Louis Pasteur said, “Chance favors the prepared mind.”

At the same time, common carrier telephone regulation remained in place—and in some ways was made stronger by the 1996 Act. The unbundling provision—the requirement that the historical local phone companies make their parts of their networks available to local competitors at government set prices—was controversial and proved difficult to implement. The problem was this: the Act was now repealing the incumbent phone companies’ life-long monopoly franchise, and theirs was the only telephone infrastructure in place. The idea that cable and wireless would supplant them was not yet popular or substantiated by experience. So the Act included the unbundling requirement in order to rein in the local phone wholesale price the FCC and states established for companies’ (presumably) temporary monopoly access to the phone companies’ loops, therefore, power over local networks and to assist the was well below the actual historical cost structure interexchange carriers like AT&T and MCI that of the incumbents who, predictably, resisted lacked local network facilities but who faced the strenuously. prospect of competition from the Bells in long distance. And so, the 1996 Act left behind a trail of controversies that have made up most of the The pricing formula the FCC established telecommunications policy agenda since then. to implement unbundling became a major First, where is the line to be drawn between issue and influence on the development of information services and telecommunications, telecommunications. As opposed to using prices and would Internet services continue to be based on actual historical or embedded costs, past considered deregulated information services depreciation patterns, and the like, it chose to or regulated common carrier services? Second, use a forward-looking method based on what a where—if at all—should “unbundling” and sharing theoretical future provider would charge, a provider of facilities be required? Third, when would we that had built the most efficient contemporary decide that competition had developed to the networks using the best available technology. This point that market forces could be relied upon to standard was known as (yet another oppressive guide it, in lieu of the public utility perspective acronym) TELRIC—total element long-run behind these regulations, and the ultimate goal incremental cost. The problem was that TELRIC (as expressed in the preamble of the 1996 Act) of was about what a prospective, efficient, new a “pro-competitive, deregulatory national policy competitor would experience, which was far from framework” achieved? These questions defined the the actual situation of the incumbent telephone broadband policy agenda for much of the coming companies, who had big, expensive legacy systems decade. left over from their previous life as heavily regulated entities. The long-run, future-oriented

7 Policy Memo Progressive Policy Institute

The World According to Kennard because the 1996 Act, in its focu-s on competition “I want to create an oasis from regulation in the within telephone and cable silos, created two broadband world, so that any company, using different playing fields for the two technologies. any technology, will have incentives to deploy Cable systems were heading away from the most broadband in an unregulated or significantly onerous elements of regulation of their business, deregulated environment. And that does not and they were investing heavily. They upgraded mean just cable companies. We must have their networks to handle digital signals after the fast and ubiquitous deployment of broadband 1996 Act was made law, and their systems, where services and that will only happen if every available, provided a very high quality of service sector of the industry has incentives to provide to residential users. By the time Kennard spoke in it: wireline, wireless and cable.” 1999, the cable companies were the cutting edge of providing broadband, deploying many of the first Remarks of William E. Kennard before the United high-speed Internet access services in the nation. States Telecom Association Annual Convention, October 18, 1999 The response to the cable-based broadband offerings by the telephone systems was the more FCC Chairman Bill Kennard’s statement, made vigorous deployment of DSL (Digital Subscriber three years after the 1996 Act, was both a Line), a technology that allowed digital data summary of what had happened since the 1996 Act services to be provided on the same copper wire and a statement of what was expected to happen as voice by using a higher frequency along the line. soon thereafter. Phone companies began introducing DSL by the end of the decade, driven by consumer demands Chairman Kennard’s vision of the long-term and in direct response to cable’s Internet offerings, future of the Internet, once seen as an over-the- but they generally lagged behind their cable horizon issue, was now explicitly one of facilities- competitors. based, inter-modal competition, and the center of that competition was no longer separate markets The slow response by phone companies to the new for broadband and telephone (and potentially interest in the high-speed Internet was probably video), but their convergence. And it was a also driven by the imbalances found in the 1996 deregulated future—an “oasis” from regulation Act. A cable company, as a provider of information would be created in order to give “every sector of services under the 1996 Act, had something closer the industry” the incentives to invest in it. to the customary right to deploy its investment as it saw fit, although it still had a variety of regulatory mandates to fulfill. But the 1996 Act limited the An “oasis” from incentives of telephone companies to invest in regulation would be Internet improvements by imposing regulatory created in order to give “unbundling requirements,” which required telephone companies to share their copper phone “every sector of the lines at the TELRIC price standard. In essence, industry” the incentives anyone who wanted to pursue it could lease access to invest in it. from the phone company and enter the business in competition with the phone company at wholesale prices, whether used for voice communication or broadband connection service. This put the phone But perhaps the most telling part of this quote companies in a Catch-22. They had to sell the is that the deployment of broadband “does not most important component of DSL service—the mean just cable companies.” That reflected both network that conveys it—to their competitors. But reality and Kennard’s vision. It reflected reality if they made the significant investments needed

8 Policy Memo Progressive Policy Institute

to offer a better product—for example, by laying induced DSL competitors, who had built optical fiber for the “last mile” from the network broadband businesses based on unbundled to the home—they could find themselves forced wholesale access to the Baby Bell’s physical to resell that improved capacity to competitors as networks. While still a competitor in today’s well. As a result, they had little incentive to make market, Earthlink’s revenues peaked in 2003. new investments, and were being outcompeted Covad was valued at $10 billion in 1999 and close due to regulation of the old ones. They were to nothing a few years later. These meteoric rises frozen in place, (a situation that now confronts the and precipitous falls reflected the nature of these European Union, which uses a similar system and companies—they made no investments in their is falling behind the United States). own infrastructure, which limited their ability to innovate, and their profitability was a product of The freezing of the phone companies in this regulatory privilege, not their own inherent ability fashion slowed the nation’s overall rate of to create value. investment in high-speed Internet via fiber and DSL. Critics of the industry often seize upon By the end of the 1990’s, these dynamics this performance and use it as an argument were coming to dominate the debate. Cable to impose yet-stricter regulation of Internet was winning the race to provide high-speed providers. But, as described below, once the FCC broadband, and telephone competitors were at a eliminated unbundling requirements for new disadvantage due to the unbundling requirement. fiber networks in 2003, and the courts set aside The vision laid out by Chairman Kennard in line-sharing unbundling more generally in 2004, the above quote was widely understood and had telco investment in fiber, either directly to the bipartisan support—in that year, he went further, home (Verizon) or to neighborhood nodes (AT&T), and laid out a visionary plan for the FCC’s role began quickly and earnestly. Unconstrained by in a competitive, post-regulatory environment.3 mandatory unbundling, cable companies had But the vision still had to be reconciled with the already responded with new versions of their questions left behind by the 1996 Act—where was DOCSIS high-speed data standard that met or the line between an information service and a exceeded the speeds of their rivals who, now telecommunications service to be drawn, and was similarly treated, responded in turn. As a result, such a line an eternal fact of life, or a creature of the United States, driven by new competition the technologies available at the time? Was the between the telcos, cable companies, and wireless transitional device of “unbundling” really just a providers, now lags only Japan among the transition? To whom would it apply, and when G-7 nations, and only highly urbanized Japan might it end? The next decade would answer many and Korea (nations with very high population of these questions. densities, a key cost consideration in providing wired service) in connection speeds. So industry The Rise of Information Services critics found themselves in the awkward position As technology progressed on many fronts, the of advocating unbundling as the policy needed convergence of the many sources of broadband to improve U.S. broadband’s performance, even connectivity began in earnest, and true though that specific policy was directly responsible competition was the result. While this was an for the lag in U.S. high-speed broadband adoption. unambiguously good thing, it left an important The industry sprang forward, both in absolute issue unresolved—was broadband access in speed and in international rankings, only once all cases an “information service” or was it the policies the industry’s critics advocated were a “telecommunications service” that should be abandoned. subjected to the same type of common carriage utility style regulation that applied to voice The inroads made by cable during this period also communications? hurt both the telcos and the telcos’ regulation-

9 Policy Memo Progressive Policy Institute

While the original intent of the Clinton that would make it subject to common carrier Administration and the Congress seemed regulation. clear, it was not until after a series of regulatory pronouncements and court decrees that common Then on August 21, 2003, the FCC released carriage was taken off the table for all services a “triennial review order” that eliminated the except local phone networks.4 And, through unbundling requirement for fiber-to-the-home these decisions, the regulators and courts made broadband capacity. (It also eliminated the important statements about “information services” unbundling requirement for “hybrid” fiber and and “ services,” and the copper loops, but allowed that states may find networks over which they were provided. it necessary to re-impose those measures when some local competitors would be “impaired” by disallowing them wholesale access.) Verizon almost The FCC under Chairman immediately began formulating and implementing plans to deploy fiber into neighborhoods and Michael Powell ruled that ultimately, to residential premises now that the Internet access provided by disincentive to do so had been eliminated.

cable companies should be The decision to liberate fiber was not universally classified as an “information acclaimed. One FCC Commissioner argued service” and not a that “today’s decision chokes off competition in broadband. Consumers, innovation, entrepreneurs “telecommunications and the Internet itself are going to suffer. . . . This service”. is not a brave new world of broadband, but simply the old system of local monopoly dressed up in a digital cloak.”5 In retrospect, it is difficult to In 2002, after a regulatory process that spanned defend that prediction. The introduction of fiber- two Administrations of both parties, the FCC to-the-home was both spurred by cable’s entry into under Chairman Michael Powell ruled that broadband and, in turn, produced a competitive Internet access provided by cable companies response from the cable industry that has should be classified as an “information service” improved the U.S. absolute and relative broadband and not a “telecommunications service,” performance dramatically, demonstrating the value removing the specter of wholesale imposition of of facilities-based competition. Thus, deregulation of a wide range of common carrier requirements, “telco” broadband actually spurred investment by including unbundling. Opponents had claimed cable, telco, and wireless competitors, a cycle that that, even though cable modem service brought continues today, as Advanced DSL and DOCSIS consumers the entirety of the Internet (and that 3.1 are readied for market, and 4G LTE becomes was why consumers purchased the service) the ubiquitous. It is hard to imagine that we would underlying transmission was itself the sale of have been better off if continued unbundling telecommunications to the public and thus should requirements had prohibited fiber deployment, be classified as telecommunications (i.e. common much as it would be hard to argue that we would carrier) services. But, in a March 14, 2002, have been better off if providers of cable modems Declaratory Ruling, Chairman Powell announced were forced to open their facilities to competitors. that the FCC had “settled a debate over the regulatory classification of cable modem service Over time, the Courts provided clarity on the and launched a proceeding to examine the proper nature of the mandates applied to different groups regulatory treatment of this service” and that cable of providers. For example, the FCC’s Triennial modem service did not contain a separate and Review Order that eliminated unbundling for distinct “telecommunications service” offering fiber loops did not provide the same treatment

10 Policy Memo Progressive Policy Institute

for hybrid loops of fiber and copper under some of the difference between telecommunications circumstances. But when industry groups appealed and information networks discussed above have that part of the order, the U.S. Court of Appeals centered on their essential character. But as the for the District of Columbia Circuit in March 2004, last decade progressed, the debate soon moved essentially rescinded any unbundling requirement on to their engineering and logistical nature. save for a minimal level of access—64 kbps, In particular, two policy proposals moved to enough for voice service, reflecting the original the fore—“net neutrality” and the regulation of intent (to preserve telecommunications services) network interconnection—each of which seeks of the 1996 Act. And in June 2005, the Supreme to have broadband networks behave like their Court ruled in what is known as the Brand X case.6 telephone antecedents. These have become In that case, a small Internet service provider in perhaps the most loudly debated, if not most Santa Monica, California, argued that the cable important, aspect of policy toward the Internet modem ruling of 2002 was in error, that the data today. communication portion of a cable modem service was a telecommunications service, and that it was “Neutrality” first: The telephone system of a therefore subject to common carrier regulations generation ago was “dumb”—it did little more than requiring (among other things) cable companies to set up an electrical circuit between a calling to unbundle it for competitors. But the Court party and a called party. In the analog world, that ruled that the FCC’s determination that cable voice signal was a continuous wave carried without broadband was an “information service” was interruption from one destination to another—that reasonable and could stand. was the only way the system could work. If the system was overloaded, the caller got a busy signal After Brand X (and follow on proceedings at and came back later. the FCC), the basic questions regarding the application of the 1996 Act structure to emerging By contrast, under the protocols that now govern Internet technologies and innovations appeared not just voice transmission but all data, image, resolved (although controversies remained video, or other streams on the digital Internet, as the FCC worked through the implications all messages are broken down into “packets” that of the decision for DSL, wireless broadband, find their way across the Internet individually and and other services). The Courts and the FCC are reassembled when they arrive, a technology had converged around one view—that simply that lies behind the increases in speed, network because an “information service” contained a utilization, quality, and declines in cost of the “telecommunications” component that did not past decades. That means that networks can be make it a “telecommunications service” subject managed to optimize their performance—they’re to public utility common carrier regulation. This not “dumb,” as was the old phone system. was the logical interpretation of the 1996 Act’s provisions in light of dramatic technological This difference emerged as a policy issue in change, which was ratified immediately by the mid-2000s. In 2005, for example, the capital markets that poured vast investment into FCC fined Madison River, a North Carolina companies and technologies when it became clear telecommunications company, for blocking the common carriage regime (and, therefore, the Vonage’s VoIP traffic, which potentially competed possibility of unbundling) would not apply. In this with Madison River’s own telephone service. context, at least, the policy intents of the 1996 Act The question of limiting certain types of traffic and of Kennard’s 1999 vision had finally been became more nuanced in 2007, when Comcast achieved. began to manage traffic that used BitTorrent, a program used for very high volume applications Net Neutrality and Network Management like peer-sharing music, video, or other large The various regulatory and legal considerations files. To some, the decisions to limit BitTorrent

1111 Policy Memo Progressive Policy Institute

during periods of Internet congestion could be ignores the reality that different Internet traffic analogized to a grocery store deciding how many can have different needs for speed and reliability. registers to open and how many lanes should be Because the broadband network has intelligence, devoted to those customers with a limited number it could readily offer different levels of service, items. But critics raised the possibility of Internet allowing some content to move more rapidly (for providers editorially managing the traffic on their instance, by keeping their packets together instead system—deciding what traffic would move and at of distributing them—thus eliminating “buffering” what speed based on what the providers “liked” as they are reassembled). This would be one way or “didn’t like.” Perhaps this would be editorial to resolve congestion, and the ability to use a control, perhaps they would quash traffic that “premium” service would greatly assist innovations competed with their own (for example, blocking that require such an unbuffered connection, Netflix because it potentially competes with a such as remote medicine, distance learning, provider’s video-on-demand). These advocates entertainment and gaming, and the like. ultimately petitioned the FCC to prohibit the practice as applying to BitTorrent, and the Net neutrality opponents argue that allowing episode became an emblem for the policy of “net content to pay for the higher quality of their neutrality” (which postulates that all traffic should signal (much as we allow households to pay for the move across the Internet under identical terms higher speed of their connection) could reduce and conditions, much as telephone calls did under the share of network costs borne by end users. An the regulated, analog voice system). In essence, analogy can be made to a newspaper—it charges neutrality advocates wanted the broadband advertisers to reach its readers, and readers to network to act the way the old phone system see its advertisements—it is a “two-sided market.” did. Moreover, they argued that the Internet If a newspaper were prohibited from charging “always worked that way,” and that this equality of advertisers for space, the cost of the newspaper treatment of all data was essential to its character. to readers would be much higher. Not charging content (websites) for premium access, therefore, potentially increases costs to users, who end The idea that all traffic up subsidizing content, including content they must be treated equally may not use and that others in the market would sounds democratic, but readily support. can be costly and This debate over neutrality rages today. The inefficient in practice, and purpose of this paper is not to resolve it, but to ignores the reality that place it in historical context. different Internet traffic The debate first flickered to life in D.C. policy can have different needs circles when Chairman Powell spoke on for speed and reliability. “Preserving Internet Freedom” at the February 2004 Silicon Flatirons conference. The following year, the issue was joined in earnest when the FCC issued its “Internet Policy Statement,” an To neutrality’s detractors, these arguments missed attempt to identify general principles that might the point. The Internet that “always worked that guide the management of network traffic. That way” was one that delivered files and mail to statement said that consumers were entitled to: users, not data-intense video files and other large access the lawful Internet content of their choice; claimants to bandwidth. Moreover, the idea that all run applications and services of their choice traffic must be treated equally sounds democratic, subject to the needs of law enforcement; connect but can be costly and inefficient in practice, and to any device that does not harm the network;

12 Policy Memo Progressive Policy Institute

and a competitive market for all components of had to be mandatory because otherwise any of the the Internet “package”—connectivity, devices, system’s sanctioned monopolists (in either long services, applications, and the like. Although the distance or local service) could try to jack up the Commission did not adopt rules in this regard, it price of interconnection to the other by holding said it would incorporate those principles into its the system hostage, and all heck would break lose. ongoing policymaking activities. But the Internet works differently. Websites But at the same time, it recognized that those generally take their content (either on their own principles were subject to “reasonable network or through a contacted service provider) to one management,” which left situations like the of many “backbone” networks that make up Comcast/BitTorrent instance in a regulatory the Internet, which in turn take it to the local grey area. The FCC ordered Comcast to make networks in your neighborhood. There are many its network management practices transparent. of these backbone providers—not just ISPs, but But what gave the FCC the right to regulate the also such more specialized companies as Cogent, way information services were managed? If a Tata, and others—and they move data across telephone company in a prior age had created the network to minimize cost and maximize a system that allowed some users to bypass a speed and efficiency through a flexible system of busy signal, the FCC would have prohibited it, arrangements. If these backbone providers billed as part of its mandate to regulate a sanctioned each other every time a message jumped from monopoly. But broadband (an information, one of their lines to the others, they’d go nuts not telecommunications, service) exists in a processing the transactions. So, instead, when the competitive environment funded by investors. two parties are roughly the same size in terms of What gave the FCC the authority to regulate volume of data transferred, they set up a peering network management practices in that world the relationship, which says that they will trade data way it did in the first? without billing so long as the volume moving both ways is “roughly” in balance. If their traffic falls Comcast took the FCC to court over the BitTorrent out of balance, they settle up and figure out how order and, in 2010, the U.S. Court of Appeals for to manage the imbalance. When two networks’ the D.C. Circuit found that the FCC had failed to traffic exchanges are dramatically different in show that its decision was reasonably related to size, other “settlement based” or commercial its statutory authority. Taking this authority unto arrangements allow this exchange to occur. itself, the Court found, would “virtually free the Commission from its congressional tether,” giving “Peering” and other arrangements, therefore, are the FCC the ability to impose regulations on the Internet’s effective free-market substitute Internet service providers that were not based on for mandatory and regulated interconnection, a Congress’ expressed intent. A second and more system that makes the competitive backbone basic legal challenge to the FCC’s ability to impose “market” work just as the old system made the neutrality on the Internet will be discussed below. prior-day, regulated monopolies work. The system has been tested, several times, most A second regulatory proposal would recreate publicly in an instance involving Level 3 and the old phone system’s practices regarding Comcast. Level 3 and Comcast had a settlement- “interconnection.” The old phone system, as free peering agreement, one that required that mentioned above, had both long distance and traffic exchanges be roughly in balance, per the local carriers. Regulation required all of them to companies’ respective published peering policies. interconnect at a specified “price.” “Price” is set in Then Level 3 entered into an arrangement to carry quotes because what was really being set was the huge amounts of Netflix’s movies on its network, way in which revenues would be divided among which resulted in the balance of traffic between the system’s participants. And interconnection Level 3 and Comcast falling far out of balance.

13 Policy Memo Progressive Policy Institute

Comcast asked to move to “settlement based” prohibit differentiated service (so long, of course, peering but Level 3 refused to pay and escalated as it didn’t violate antitrust or other consumer its argument to the FCC, insisting that Comcast’s protection statutes), or regulate interconnection, demand that they renegotiate their peering was a because of the difference between information and violation of an “open” Internet and the FCC’s oft- telecommunications services, the same distinction stated principle of “net neutrality.” FCC Chairman the FCC cited when freeing the latter from Julius Genachowski, in remarks before Congress, common carriage requirements. leaned toward Comcast’s view, arguing that peering disputes were private business matters, The Court’s decision in this case was delivered and expressed the hope that companies like Level in January 2014. While it acknowledged the 3 and Comcast could work out their differences FCC’s role in promoting the Internet, it found as was routinely done in the marketplace. (The that the FCC, having classified broadband ISPs companies did in fact work out an agreement, as information service providers, could not without regulatory intervention.) impose common carrier-type regulations on them. In response, many advocates for neutrality A similar dispute is now underway with Cogent— and similar regulation now suggest that the yet another conveyor of Netflix content—and FCC identify (or “reclassify”) the Internet as Verizon. (Note that video plays a role in both these a telecommunications service rather than examples.) And, again, advocates for regulating an information service, thereby undoing the network interconnections see this dispute over distinction first made in the law in the 1996 Act, traffic exchange among networks—a dispute and undoing the Clinton Administration’s guiding resolved by regulation under the rules of the old intent. phone system—as an example of the need for regulation of the new broadband system, even What Are the Lessons? though the vast bulk of Internet traffic moves Different analysts will take different lessons from through these market-based arrangements without this history. This review sees three of primary incident. importance. They are:

Thus, the debate over regulation of the Internet • information services and telecommunications has moved from the issue of common carriage services really are different; to the issue of network management. But, • investment and capital flow to where regulation paradoxically, the resolution of the second (or the absence thereof) encourages them to question could lead us back to the first. In flow; and December 2010, the FCC issued its “Open Internet • technology, business models, and consumer Order,” which prohibits Internet service providers behaviors change and, as they change, the from “discriminating” against any legal content meaning and effect of different regulatory among other requirements, subject to reasonable proposals change as well. network management requirements. This is often assumed to prohibit ISPs from offering “tiered” The rest of this section examines these three services to Internet content providers, although propositions. it may be argued that allowing content providers to pick a service tier at posted prices is no more The 1996 Act codified the distinction between discriminatory than letting consumers choose information services and telecommunications among “good,” “better,” and “best” from Sears. services, and expressly limited common carrier regulation to telecommunications. The Internet Verizon, in response, appealed the Open Internet has flourished as this distinction was put into Order in the courts, claiming that the FCC lacked practice and, critically, common carriage status the authority to impose the neutrality requirement, and unbundling requirements were either lifted

14 Policy Memo Progressive Policy Institute

or were never placed on the various information would have been outrageous. Imagine, for instance, service providers that offer high-speed broadband. that we applied unbundling to peanut butter, and Peter Pan had the legal right to use Skippy’s manufacturing facilities at a price determined The 1996 Act codified by what Peter Pan’s processing costs would be if they had built a new, state-of-the-art plant (which the distinction between they did not and had no intention to do). That information services and is exactly what unbundling does. The 1996 Act, telecommunications therefore, imposed common carriage in a narrow area to correct for a historical fact, but it spared services, and expressly new forms of innovation and investment from this limited common carrier burden.

regulation to The differences between telecommunications telecommunications. services and information services recognized by the 1996 Act have been extended significantly since then, in part because of the way the 1996 Act One of the intentions of the 1996 Act was to treated them, and in part because of technological preserve the commitment made to the public in progress. The old phone network addressed by the the 1934 Act—that they would have access to local 1996 Act was built by sanctioned monopolies that telephone service at affordable rates, including traded the market power the government ceded emergency services such as 911. Cable was still to them for a regulated return and the ability to subject to a variety of regulatory mandates—its invest in facilities with minimal risk. But Internet basic tier was subject to price regulation, and it access has been provided by competing private was still subject to certain carriage and tiering parties who have risked their own investments— requirements—but the Act asserted a basic over a trillion dollars’ worth since the 1996 Act— difference between telephony and cable television, without government guarantees. This extends data communication, or other information the original basis for the distinction between flows, as had past FCC regulatory practice and the two systems—information services were the perspective of the Clinton Administration. financed by risk capital, in contrast to their legacy The distinction made in the 1996 Act between telecommunications counterpart. telecommunications and information services reflected genuine and significant differences in The system of sanctioned monopolies also gave the nature of the two services—differences in rise to mandatory interconnection and settlement functionality, in engineering characteristics, in policy regarding shared revenue (for instance, the their potential for innovation and improvement. division of dial tone revenues to local versus long The differences and distinctions between the distance calling). The system needed mandatory two made in the 1996 Act were not political or interconnection to work, and required a method semantic, but real. for allocating revenues and costs among long distance and local traffic (which essentially The distinction between the two types of services determined what the price of that interconnection was made so that there would be a guideline was) to determine profits as well as to subsidize for determining where regulation and common many local services. The Internet, in contrast, carriage status were needed. It was imposed on is a “network of networks” comprised of many providers of telephone services because these had different companies’ facilities. As opposed to their been local monopolies and, therefore, had the only regulated predecessors, these companies are in the telephone infrastructure in place. Had this not business of interconnecting. Occasional cases such been the case, the imposition of common carriage as Level 3 and Comcast, or Cogent and Verizon

15 Policy Memo Progressive Policy Institute

(or Cogent and Sprint in 2008, when Cogent had why those features have been reciprocal sides of again asserted that free “peering” did not require the basic regulated telephone bargain. But once balanced traffic loads), are the rare and usually opportunities appeared outside the regulated histrionic exceptions to a system that works well system, money poured into them. The MFJ of 1982 every day. And there is a serious risk that undue opened the gates to aggressive investment and regulation in this well-working area may only competition in long distance markets. Investment generate or incent such disputes, rather than poured into cable television in the 1980s once eliminate them. the 1984 Cable Act reduced regulation and standardized it across the national market. It then backed off after the 1992 Act mandated Investment goes to where new price regulations. But when the 1996 Act regulation allows and rolled back aggressive cable price regulation, encourages it to go. cable began to attract major investment again, not just for television service, but to implement the DOCSIS standard that brought the United Moreover, as mentioned earlier, the old phone States into the world of wired broadband. And the system was a “dumb” system that connected absence of regulation also supported the 1990s user to user without any control over their boom in fiber Internet backbone, by allowing communication absent breaking the circuit. And interconnection to be negotiated by private parties the signals that it carried were homogeneous— in a competitive market. Fiber is being installed phone calls varied only in how long they wanted at a much faster rate in the United States than it their circuit to remain open. But because of is in Europe; what would happen to investment packet switching and the digitization of all forms in fiber Internet backbone in the United States if of information—voice, data, images, video, or it were to suddenly and unanticipatedly be made what-have-you—the traffic carried by the Internet subject to federal mandates regarding carriage and varies widely, meaning that the cost to the network retransmission and price oversight? and the implications for network management of different signals can be very different. This means The same dynamics can be found regarding that different types of traffic impose different wireless, DSL, and fiber-based systems. Wireless costs on the system (through congestion) and that systems were freed from most price regulation the economic value of traffic to the user can vary in the 1993 Omnibus Budget Reconciliation widely (for example, an interbank transaction Act, and wireless broadband was subsequently versus your e-mail to your aunt). And if users classified as an information service, which has wanted more “quality” in the regulated phone allowed investment in those systems to remain world, it happened outside the regulated system— strong ever since, rising from under $20 billion big users bought private networks. An important annually in 1996 to about $35-40 billion difference between the two systems, therefore, annually for most of the past decade.7 The rise of is that today’s information systems allow the DSL access over a decade ago, in contrast, was consumer to match price and quality as she or he wholly a matter of giving competitors access to sees fit if we allow them to do so. existing infrastructure and, as a result, little new infrastructure was built. But once new fiber A second overall lesson learned from this and hybrid system were taken out from under experience is that investment goes to where the burden of common carriage and threats of regulation allows and encourages it to go. unbundling, investment in them was sudden and When price regulation or common carriage is sizable, triggering a new round of competition that imposed, the only way to secure investment is has brought the United States from 22nd to 9th, to guarantee a return to it, with the predictable and rising, in international rankings. effects on innovation and expansion. That is

16 Policy Memo Progressive Policy Institute

Thus, investment goes where regulation guides information systems was the network itself—they it by making it either welcome or unwelcome. were “network centric.” The old phone system And both common carriage and a prohibition represented that view—the only purpose for the on service tiering have this characteristic; they equipment you bought was to reach the network. throttle the flow of capital into the sector and are But after the explosion in devices triggered by therefore implemented at a potentially great cost, the iPhone, and the proliferation of “apps” and particularly in light of the stated policy goal of services the iPhone has allowed us to imagine, expanding and improving the Internet. the model of how broadband creates value has changed dramatically, as first discussed in a A third and final lesson of this experience seminal paper on “the consumer value circle” by is that technology constantly changes, and Jonathan Sallet.8 regulations and procedures that are premised on a set of technological “facts” may be severely The “value circle” argument sees the iPhone challenged when those facts change. The 1996 as turning the idea that all value resides in the Act demonstrates this problem. While it may have network on its head. Where devices were once mostly anticipated intra-model competition, inter- simple attachments to the network (such as the modal competition has actually blossomed as well. black handsets that populate Moreover, broadband was a far-away-future when film noir, or the monopoly era’s most compelling the 1996 Act was drafted, and wireless broadband “innovation,” the Princess phone), the network was simply not yet imagined, although it is perhaps has increasingly become merely the stage on the fastest growing component of competitive broadband today and the one in which the United States is an unambiguous world leader. This change in market structure, driven by both Technological change has affected broadband regulation in other ways. The ability to compress innovation in both video, music, and other types of content has made devices and Internet traffic far more varied, unlike the voice connections, has signals carried by the analog telephone networks of a generation ago, and unlike the simple file changed the market, but sharing that led to the creation of ARPANet and regulation has yet to other precursors of the modern Internet. This legitimately raises the question of whether to catch up. allow service differentiation, which was arguably irrelevant under the old phone system or the Internet’s precursors, but is very relevant given the which the device does spectacular things. And Internet’s great capabilities and the broad range of as post-iPhone devices have grown in power and users and users it serves. sophistication, an entire new industry emerged in applications that compete with the devices But perhaps the most subtle, yet pivotal, for consumer allegiance as much as the devices technological change that challenges our ideas themselves compete with the service providers about Internet regulation is the rise of devices, that host them. Rather than an edifice that rests applications, and services, a change triggered by on the signal of Internet providers, the broadband the introduction of the iPhone. The FCC’s various experience is now an integrated proposition in statements about the management of the Internet, which signal, devices, content, and applications going back to the Internet Policy Statement in all compete to be the organizing framework for 2005, have all been based on the view that the the consumer’s experience—the broadband value choke point in both telecommunications and proposition. Each is now a platform in its own

17 Policy Memo Progressive Policy Institute

right and competes to be the part of the experience on a treadmill to provide better and better signals to which the consumer bears allegiance, whether so that Apple or Android phones, or apps such as it’s “Comcast has the fastest connections,” or “the Facebook or Twitter, or services such as Google, iPhone is better than all the others,” or “Google is can deliver more value to the consumer by using the best way to see the world’s information,” or “to them, growing profitably in the process. me, the Internet is Facebook and Twitter.” Each is competing for a larger slice of the “pie” of value So the new business dynamics in the mobile the consumer assigns to the integrated broadband broadband world are these; mobile ISPs invest experience. and improve their service, which allows the functionality of devices and the desirability of apps Moreover, by the traditional measures, many and services to increase and, paradoxically, reduces of these other components of the integrated the market power of carriers by commoditizing broadband proposition are even more them in the eyes of the consumer. In fact, they are “concentrated” than the provision of signal. The not “commodities”—their ongoing innovation and leading operating systems for mobile phones hold a improvements speak to that. But their ability to far larger share of their market than do the largest gain market power is limited in the face of these ISPs in theirs. The “concentration” of social media, new dynamics. Internet provision, because of these or Internet search, or other key applications or dynamics, is rapidly becoming a classic example of services are also just as, if not more, concentrated the traditional economic definition of an “epochal than broadband providers. invention”—more money is made using it than providing it. And yet regulation is still organized This change in market structure, driven by both around the idea that the network itself realizes all innovation in devices and connections, has of the value created by the broadband experience, changed the market, but regulation has yet to much as it realized all of the value in the phone catch up. The Internet Policy Statement and other system of a generation—if not a lifetime—ago. “network-centric” policies did not anticipate that these other value-creating elements force ISPs Conclusion to compete just as another ISP would, if only We are now in the middle of a heated debate over because they are contesting shares of the total Internet regulation, with the possibility of an even broadband experience. Consider this example: as more feverish one now that the courts have ruled wireless signal providers offer stronger and more in Verizon vs. FCC. Many proposals have been made reliable connections, the capabilities of the devices to regulate Internet providers that have their roots they support grow. Look at the iPhone’s voice- in the regulatory history of the telephone system, recognition capabilities. Voice recognition, as but while the history of that system and the anybody who has ever talked to a customer service transition away from it is now commonly known, “agent” knows, has been around for a long time. the way in which policy proposals have interacted But it only became practical to use on a mobile with the system over time is not. As this paper device once connections became good enough to seeks to make clear, many of the proposals being support real-time communication between your debated as “new” and “necessary” are neither— phone and the voice-recognition hardware in and actually originated in a dramatically different “the cloud.” So once real-time connectivity was place and time and set of circumstances that made possible, thanks to the investments and bears no relationship to the marketplace of today. innovations of the mobile carriers, Apple provided Moreover, history makes clear that regulation an innovation that used that connectivity. That influences investment and is often undone by made the iPhone more valuable, but did little to technological change. Any view on regulatory improve the market power of the carriers who proposals for the broadband world should respect made the innovation possible. Instead, ISPs are that history and the lessons it has to teach.

18 Policy Memo Progressive Policy Institute

Endnotes 1. See Ehrlich, the State of the Broadband Internet, Progressive Policy Institute, April 2014 (forthcoming). 2. Many other authors have investigated this history in more time and detail than allowed here. One such work is Jonathan Nuechterlein and Philip J. Weiser, Digital Crossroads: American Telecommunications Policy in the Internet Age, (2nd edition, MIT Press 2013). 3. Federal Communications Commission, “Strategic Plan: A New FCC for the 21st Century,” August 1999: http://transition.fcc. gov/21st_century/draft_strategic_plan.pdf. 4. That is, in contrast to the situation that existed when the Congress passed the 1996 Act, a majority of the country now relies on other than copper loop for its local phone calls—cable has replaced the phone company in many of these applications, and a steadily rising share of households use only mobile phones. 5. FCC Press Office, “FCC Commissioner Michael J. Copps Reacts to Release of Long-Awaited Triennial Review Decision,” August 21 2003: http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-238137A1.pdf. 6. National Cable & Telecommunications Assn. V. Brand X Internet Services. 345 F.3d (U.S. 967 2005). http://www.law.cornell.edu/ supct/html/04-277.ZS.html. 7. US Telecom, “Broadband Investment,” 2013: http://www.ustelecom.org/broadband-industry/broadband-industry-stats/investment. 8. Jonathan Sallet, “The Creation of Value: The Broadband Value Circle and Evolving Market,” Structures Social Science Research Network, April 4, 2011: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1821267.

Photos courtesy of the William J. Clinton Library.

19 About the Progressive Policy Institute

The Progressive Policy Institute (PPI) is an independent © 2014 research institution that seeks to define and promote a new Progressive Policy Institute progressive politics in the 21st century. Through research, All rights reserved. policy analysis and dialogue, PPI challenges the status quo Progressive Policy Institute and advocates for radical policy solutions. 1101 14th St. NW Suite 1250 Washington, DC 20005

Tel 202.525.3926 Fax 202.525.3941 Email [email protected] www.progressivepolicy.org